Neurology · Medical Literature
Placebo-controlled three-armed pilot trial of Escitalopram or Nortriptyline for depressive symptoms in Parkinson's disease (ADepT-PD).
Abstract
Pilot randomised-controlled trial of nortriptyline and escitalopram compared to placebo for depressive symptoms in people with Parkinson's disease (PD). Participants with PD and depressive symptoms received either nortriptyline, escitalopram or placebo with assessments at baseline and 8 weeks. Feasibility of the full trial was assessed using recruitment rate, loss to follow up before the 8-week primary endpoint, adherence to trial medication and rate of clinically significant adverse reactions. Intended efficacy outcomes included the BDI-II, Patient Health Questionnaire (PHQ-9), Parkinson Anxiety Scale (PAS), MDS-UPDRS and adverse effects. The aim was to recruit 46 participants to determine feasibility of the full trial. 52 participants were recruited from 24 NHS sites and randomised to nortriptyline (n = 16), escitalopram (n = 17) or placebo (n = 19). However, despite multiple strategies, recruitment took two years, and the study therefore did not reach its feasibility aim. Exploratory analyses showed that at 8 weeks depressive symptoms decreased significantly in all three arms. There was no difference in BDI-II score changes between the nortriptyline or the escitalopram when compared to the placebo arm, but PHQ-9 scores showed a greater reduction compared to placebo with nortriptyline (-3.4, 95% CI -5.75 to -0.95) but not with escitalopram (-1.2, 95% CI -3.54 to -1.17). Persistent Anxiety Scale scores showed a similar pattern (nortriptyline: -2.8, 95% CI -5.5 to -0.01; escitalopram: -2.2, 95% CI -4.9 to 0.05). As the study did not reach its primary aim of feasibility due to recruitment difficulties it did not continue to the full trial. The exploratory results are in keeping with previous studies suggesting a beneficial effect of nortriptyline on symptoms of depression and persistent anxiety. The small sample size however limits conclusions on efficacy and a beneficial effect of escitalopram is not excluded. STUDY DATASET IS AVAILABLE ON CLINICALTRIALS.GOV ID: NCT03652870, registered 29 August 2018.
Clinical relevance
In adults with Parkinson's disease and clinically significant depressive symptoms, is treatment with escitalopram (SSRI) or nortriptyline (TCA) more effective than placebo in reducing depressive symptoms over 8 weeks, and is a definitive full-scale trial feasible? __SNAPSHOT__ {"studyDesign":"Three-arm, placebo-controlled, double-blind, randomised controlled pilot trial with internal feasibility assessment (phase IV design); part of the larger ADepT-PD programme","setting":"Multicentre; 24 NHS sites across the United Kingdom","sampleSize":"n=52 (nortriptyline: n=16; escitalopram: n=17; placebo: n=19); target feasibility threshold was 46 participants, but this was not met within the intended timeline","population":"Adults with confirmed Parkinson's disease and clinically significant depressive symptoms (BDI-II ≥14), including subsyndromal depression, major depressive disorder, or persistent depressive disorder. Excluded: MoCA <16 or incapacity to consent, antidepressant use within 4 weeks, significant cardiac arrhythmias, severe liver failure, QTc prolongation, contraindicated concomitant medications (e.g., irreversible MAO inhibitors)","intervention":"Nortriptyline (TCA; titrated up to 100 mg/day) or escitalopram (SSRI; up to 20 mg/day) for up to 8 weeks (originally planned 12 months; truncated by funder due to recruitment failure)","comparator":"Matched placebo (double-blind, identical-appearing tablets)","primaryOutcome":"Change from baseline in BDI-II total score at 8 weeks; trial-level primary aim was feasibility (recruitment rate, loss to follow-up before 8-week endpoint, adherence to trial medication, rate of clinically significant adverse reactions)","followUp":"8 weeks (originally planned 52 weeks; long-term follow-up was discontinued when the NIHR truncated the trial in September 2022)","fundingSource":"Government (NIHR Health Technology Assessment programme, award ref: 16/145/01); additional support from Cure Parkinson's Trust for the nortriptyline motor progression substudy. No industry funding declared."}
Key findings
The trial failed to meet its primary feasibility objective: although 52 participants were recruited from 24 NHS sites (exceeding the minimum 46-participant threshold numerically), recruitment extended over two years — far beyond the intended pace — leading the NIHR to truncate the trial and preclude progression to the full 408-participant RCT. Of 1,396 screened patients, only 52 were randomised. One participant in the escitalopram group withdrew due to a serious adverse event. Completion of the 8-week primary outcome assessment was high among those who continued (17/19 in placebo completed; nortriptyline and escitalopram arms: not reported separately in available data). All three arms showed reductions in BDI-II total scores from baseline to 8 weeks, including placebo. For the primary efficacy endpoint (BDI-II change), there was no statistically significant difference between either active treatment arm and placebo. On the exploratory PHQ-9 secondary measure, nortriptyline showed a greater reduction vs. placebo (mean difference: −3.4 points; 95% CI −5.75 to −0.95), while escitalopram did not (mean difference: −1.2 points; 95% CI −3.54 to −1.17). On the Parkinson Anxiety Scale (persistent anxiety subscale), nortriptyline showed a nominally larger reduction vs. placebo (mean difference: −2.8 points; 95% CI −5.5 to −0.01), and escitalopram showed a directionally similar but non-significant trend (mean difference: −2.2 points; 95% CI −4.9 to 0.05). Notably, only a small proportion of participants reached the intended target dose at 8 weeks, with slower titration and lower maximum doses due to adverse effects. Adverse effects were described as compatible with the known profiles of each medication. No NNT/NNH can be calculated as absolute response rates were not reported in the available abstract and summary data.
Practice implications
This trial established that conducting a definitive RCT of antidepressants in PD-associated depression within the NHS is operationally very challenging: despite recruiting from 24 sites over two years, the trial could not meet its feasibility criteria and was terminated before the full-trial phase. The exploratory efficacy data should be interpreted with extreme caution given the severely underpowered sample (n=52 vs. 408 required), truncated follow-up (8 vs. 52 weeks), absence of target-dose attainment in most participants, and exploratory rather than confirmatory analytic status. The nominally significant PHQ-9 and anxiety signal for nortriptyline is hypothesis-generating only and cannot be used to change prescribing. The study did not establish clinical or cost-effectiveness of either agent. The large placebo response observed across all three arms is clinically important and consistent with prior depression trials in PD, highlighting the need for placebo control in any future work. Findings are limited to patients with sufficient cognitive capacity (MoCA ≥16), no major cardiac contraindications, and without current antidepressant use — excluding many patients encountered in routine PD neurology and geriatric practice. Only abstract and secondary NIHR report summaries were available for this structured review; the full dataset and statistical tables were not directly accessible. __LIMITATIONS__ Severely underpowered: 52 participants enrolled versus 408 required for 90% power to detect a 3-point BDI-II difference; all efficacy analyses are exploratory and should not be interpreted as confirmatory.,Truncated follow-up: the trial was reduced to 8 weeks by the funder; long-term effectiveness, durability of effect, and motor/non-motor impact over 12 months cannot be assessed.,Recruitment failure and slow accrual: only 52 of 1,396 screened patients were randomised (conversion rate ~3.7% [calculated]), reflecting extensive exclusion criteria, competing trials, staff shortages, and COVID-19 disruption — severely limiting generalisability and power.,Dose inadequacy: a small proportion of participants reached target doses due to adverse effects, meaning the efficacy analyses do not reflect full pharmacological exposure and may underestimate treatment effect.,Large placebo response: BDI-II improved in all three arms, including placebo, which may dilute detectable drug-placebo differences and reflects regression to the mean, natural history effects, or non-specific care effects.,Heterogeneous depression spectrum: inclusion of subsyndromal, major, and persistent depressive disorder without stratification makes interpretation of efficacy signals difficult.,Strict exclusion criteria (MoCA <16, cardiac contraindications): results are not generalisable to cognitively impaired or medically complex PD patients — a substantial real-world proportion.,Selective outcome reporting risk: the BDI-II primary outcome was non-significant; the positive PHQ-9 finding for nortriptyline is a secondary exploratory measure and subject to multiplicity concerns.,No long-term safety, motor progression, quality of life, or cost-effectiveness data available.,Abstract-only review limitation: full adverse event data, attrition by arm, and detailed statistical tables were not available in the abstract; this summary is based on abstract and NIHR report summaries. __BOTTOM_LINE_CATEGORY__ not_ready_for_clinical_application __BOTTOM_LINE__ ADepT-PD failed its feasibility objective due to prohibitively slow recruitment across 24 NHS sites over two years, and the NIHR terminated the trial before progression to the definitive phase; no conclusions on clinical effectiveness can be drawn from the exploratory analyses alone. Exploratory data showed no BDI-II difference between either antidepressant and placebo, with a nominally significant PHQ-9 improvement for nortriptyline only, but this finding is severely limited by underpowering, dose inadequacy, and the high placebo response. A definitive, adequately powered RCT of antidepressants in PD-associated depression remains urgently needed but must address the substantial operational barriers that caused this trial's premature termination.